Replicate
Enables users to access and manage Replicate's AI models and predictions via the Model Context Protocol, offering tools for creating, canceling, and retrieving model predictions and parameters.
What is Replicate?
Replicate is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to enables users to access and manage replicate's ai models and predictions via the model context protocol, offering tools for creating, canceling, and retrieving model predictions and parameters.
Enables users to access and manage Replicate's AI models and predictions via the Model Context Protocol, offering tools for creating, canceling, and retrieving model predictions and parameters.
This server falls under the Data Science & ML category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Enables users to access and manage Replicate's AI models and
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx replicate-mcp-serverConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Replicate
The mcp-replicate server connects your AI assistant to Replicate's cloud platform, giving it the ability to search Replicate's model library, create and monitor predictions, view generated images, and manage prediction lifecycle — all through natural language. It is useful for developers who want to run image generation, video, audio, or other ML model predictions from within their AI workflow without switching to the Replicate web UI or writing API calls manually.
Prerequisites
- Node.js 18 or later (for npx / global npm install)
- A Replicate account and API token (create at replicate.com → Account Settings → API Tokens)
- An MCP client such as Claude Desktop, Cursor, or Cline
- The REPLICATE_API_TOKEN environment variable set to your token
Get your Replicate API token
Sign in at replicate.com, go to Account Settings → API Tokens, and create a new token. Copy it — you will need it in the MCP config.
Install mcp-replicate globally (optional)
You can install the package globally so it is available as a binary, or use npx to run it without installing. The global install approach is shown here.
npm install -g mcp-replicateConfigure your MCP client
Add the replicate server to your MCP client config. Set REPLICATE_API_TOKEN to your actual token. If you used the global install, the command is 'mcp-replicate'; for npx use the npx form shown.
{
"mcpServers": {
"replicate": {
"command": "mcp-replicate",
"env": {
"REPLICATE_API_TOKEN": "r8_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
}
}
}
}Restart your MCP client
Restart Claude Desktop or your chosen client to load the new server configuration. The AI should now have access to Replicate model and prediction tools.
Test by searching for a model
Ask your AI assistant to search for an image generation model on Replicate to confirm the integration is working.
Replicate Examples
Client configuration
Claude Desktop configuration using mcp-replicate with a Replicate API token.
{
"mcpServers": {
"replicate": {
"command": "mcp-replicate",
"env": {
"REPLICATE_API_TOKEN": "r8_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
}
}
}
}Prompts to try
Use natural language to interact with Replicate models and predictions through your AI assistant.
- "Search Replicate for the best text-to-image models"
- "Run a prediction with stability-ai/stable-diffusion using the prompt 'a sunset over mountain peaks'"
- "List my recent predictions on Replicate and show their status"
- "Show me the available collections of models on Replicate"
- "Cancel the prediction with ID xyz123 on Replicate"Troubleshooting Replicate
Authentication error or 401 Unauthorized
Verify REPLICATE_API_TOKEN is set correctly in the env block of your MCP config. The token should start with r8_. You can test it with: curl -H 'Authorization: Token r8_your_token' https://api.replicate.com/v1/account
Prediction times out or stays in 'starting' state
Replicate cold-starts GPU instances for some models, which can take 30-120 seconds. Use the create_and_poll_prediction tool (or ask the AI to poll for results) rather than create_prediction alone if you want the AI to wait for completion automatically.
mcp-replicate command not found after global install
Ensure npm's global bin directory is on your PATH. Check with: npm bin -g. Add that directory to your shell's PATH, or use the npx form in the MCP config: "command": "npx", "args": ["-y", "mcp-replicate"].
Frequently Asked Questions about Replicate
What is Replicate?
Replicate is a Model Context Protocol (MCP) server that enables users to access and manage replicate's ai models and predictions via the model context protocol, offering tools for creating, canceling, and retrieving model predictions and parameters. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Replicate?
Follow the installation instructions on the Replicate GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Replicate?
Replicate works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Replicate free to use?
Yes, Replicate is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.
Replicate Alternatives — Similar Data Science & ML Servers
Looking for alternatives to Replicate? Here are other popular data science & ml servers you can use with Claude, Cursor, and VS Code.
Ultrarag
★ 5.6kA Low-Code MCP Framework for Building Complex and Innovative RAG Pipelines
RocketRide
★ 3.1k📇 🏠 - MCP server that exposes RocketRide AI pipelines as t
Aix Db
★ 2.1kAix-DB 基于 LangChain/LangGraph 框架,结合 MCP Skills 多智能体协作架构,实现自然语言到数据洞察的端到端转换。
NeMo Data Designer
★ 1.9k🎨 NeMo Data Designer: Generate high-quality synthetic data from scratch or from seed data.
PaperBanana
★ 1.7kOpen source implementation and extension of Google Research’s PaperBanana for automated academic figures, diagrams, and research visuals, expanded to new domains like slide generation.
MiniMax
★ 1.5kBridges MiniMax AI capabilities to the Model Context Protocol, enabling AI agents to perform image understanding, text-to-image generation, and speech synthesis. It provides a standardized interface for accessing MiniMax's core tools via JSON-RPC.
Browse More Data Science & ML MCP Servers
Explore all data science & ml servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Replicate in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Replicate?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.